Splitting audio with 1001fx for Whisper

How To
,
1

Hello ladies and gents,

I’m creating a service for long audio transcription. I want to use OpenAI Whisper which has a limit of 25MB per file. I’m splitting audio using 1001fx module (unfortunately only mp3, no m4a). It’s output (binary data) is used as OpenAI’s input.

Here’s an issue: OpenAI module returns an error:
[400] Unrecognized file format. Supported formats: [‘flac’, ‘m4a’, ‘mp3’, ‘mp4’, ‘mpeg’, ‘mpga’, ‘oga’, ‘ogg’, ‘wav’, ‘webm’]

JSON output from 1001fx:
[
{
“data”: “IMTBuffer(149, binary, e87d1455a4927388b58fc0863c24b30a576058fa): 7b22726573756c74223a7b2275726c223a2268747470733a2f2f70726f642d3130303166782d7075626c69632e622d63646e2e6e65742f74656d706173736574732f32303234313230352f30314a4539305451303151395142573546584a565433425642”,
“fileName”: “filename”,
“fileSize”: “1”,
“statusCode”: 200
}
]

Anyone had similar issues or an idea how to do it differently?